Not applicable.
Not Applicable.
The present invention generally relates to medical diagnostic imaging systems, and in particular relates to a method and apparatus for correcting the image offset induced by Field Effect Transistor (FET) photo-conductive effects in medical imaging systems employing solid state detectors.
X-ray imaging has long been an accepted medical diagnostic tool. X-ray imaging systems are commonly used to capture, as examples, thoracic, cervical, spinal, cranial, and abdominal images that often include information necessary for a doctor to make an accurate diagnosis. X-ray imaging systems typically include an x-ray source and an x-ray sensor. When having a thoracic x-ray image taken, for example, a patient stands with his or her chest against the x-ray sensor as an x-ray technologist positions the x-ray sensor and the x-ray source at an appropriate height. X-rays produced by the source travel through the patient""s chest, and the x-ray sensor then detects the x-ray energy generated by the source and attenuated to various degrees by different parts of the body. An associated control system obtains the detected x-ray energy from the x-ray sensor and prepares a corresponding diagnostic image on a display.
The x-ray sensor may be a conventional screen/film configuration, in which the screen converts the x-rays to light that exposes the film. The x-ray sensor may also be a solid state digital image detector. Digital detectors afford a significantly greater dynamic range than conventional screen/film configurations, typically as much as two to three times greater.
One embodiment of a solid state digital x-ray detector may be comprised of a panel of semiconductor FETs and photodiodes. The FETs and photodiodes in the panel are typically arranged in rows (scan lines) and columns (data lines). A FET controller controls the order in which the FETs are turned on and off. The FETs are typically turned on, or activated, in rows. When the FETs are turned on, charge to establish the FET channel is drawn into the FET from both the source and the drain of the transistor. The source of each FET is connected to a photodiode. The drain of each FET is connected to readout electronics via data lines. Each photodiode integrates the light signal and discharges energy in proportion to the x-rays absorbed by the detector. The gates of the FETs are connected to the FET controller. The FET controller allows signals discharged from the panel of photodiodes to be read in an orderly fashion. The readout electronics convert the signals discharged from photodiodes. The energy discharged by the photodiodes in the detector and converted by the readout electronics is used by an acquisition system to activate pixels in the displayed digital diagnostic image. The panel of FETs and photodiodes is typically scanned by row. The corresponding pixels in the digital diagnostic image are typically activated in rows.
The FETs in the x-ray detector act as switches to control the charging and discharging of the photodiodes. When a FET is open, an associated photodiode is isolated from the readout electronics and is discharged during an x-ray exposure. When the FET is closed, the photodiode is recharged to an initial charge by the readout electronics. Light is emitted by a scintillator in response to x-rays absorbed from the source. The photodiodes sense the emitted light and are partially discharged. Thus, while the FETs are open, the photodiodes retain a charge representative of the x-ray dose. When a FET is closed, a desired voltage across the photodiode is restored. The measured charge amount to re-establish the desired voltage becomes a measure of the x-ray dose integrated by the photodiode during the length of the x-ray exposure.
X-ray images may be used for many purposes. For instance, internal defects in a target object may be detected. Additionally, changes in internal structure or alignment may be determined. Furthermore, the image may show the presence or absence of objects in the target. The information gained from x-ray imaging has applications in many fields, including medicine and manufacturing.
In any imaging system, x-ray or otherwise, image quality is of primary importance. In this regard, x-ray imaging systems that use digital or solid state image detectors (xe2x80x9cdigital x-ray systemsxe2x80x9d) face certain unique difficulties. Difficulties in an digital x-ray image could include image artifacts, xe2x80x9cghost images,xe2x80x9d or distortions in the digital x-ray image. One source of difficulty faced by digital x-ray systems is the photo-conductive characteristics of semiconductor devices used in the digital x-ray systems.
Photo-conductivity is an increase in electron conductivity of a material through optical (light) excitation of electrons in the material. Photo-conductive characteristics are exhibited by the FETs used as switches in solid state x-ray detectors. Ideally, FET switches isolate the photodiode from the electronics which restore the charge to and measure the charge upon the photodiode. FETs exhibiting photo-conductive characteristics do not completely isolate the photodiode from the system, when the FETs are open. Consequently, the FETs transfer excess charge to the readout electronics. If the FETs transfer excess charge to the readout electronics, the energy subsequently discharged from the photodiodes to activate the pixels in the digital image may be affected. The unintended charge leakage through the FETs may produce artifacts or may add a non-uniform offset value to each of the pixels in the digital x-ray image, thus producing a line artifact in the image.
FETs and other materials made of amorphous silicon also exhibit a characteristic referred to as charge retention. Charge retention is a structured phenomenon and may be controlled to a certain extent. Charge retention corresponds to the phenomenon whereby not all of the charge drawn into the FET to establish a conducting channel is forced out when the FET is turned off. The retained charge leaks out of the FET over time, even after the FET is turned off, and the leaked charge from the FET adds an offset to the signal read out of the photodiodes by the x-ray control system.
The FETs in the x-ray detector exhibit charge retention characteristics when voltage is applied to the gates of the FETs to read the rows of the x-ray detector. The detector rows are generally read in a predetermined manner, sequence, and time interval. The time interval may vary between read operations for complete frames of the x-ray image. When a FET is opened after the charge on an associated photodiode is read by a charge measurement unit, the FET retains a portion of the charge. Between read operations, the charge retained by the FETs leaks from the FETs to a charge measurement unit. The amount of charge that leaks from the FETs exponentially decays over time. The next read operation occurs before the entire retention charge leaks from the FETs. Consequently, the charge measurement unit measures during each read operation an amount of charge that was stored by the FETs during the previous read operation. The charge measurement unit also reads an amount of charge that was stored by FETs that were activated in scan lines preceeding the current scan line in the current read operation.
The charge remaining on the FETs when a new read operation is initiated is referred to as the initial charge retention. The initial charge retention stored on multiple FETs, such as the FETs of a single data line, combines to form a charge retention offset for that column. The charge retention offset varies based on the rate at which rows of the x-ray detector panel are read. As the interval increases between read operations, the charge decay increases. As the panel rows are read, the charge retention offset builds to a steady state value. The steady state value for the charge retention rate represents the point at which the panel rows are read at a rate equaling the exponential decay rate of the charge on the FETs.
If the times between frames for both the offset and x-ray image are consistent, the effect of charge retention may be eliminated from the final image. In the normal process of reading a detector, the effect of retained charge may be minimized by simply subtracting the results of a xe2x80x9cdarkxe2x80x9d scan from the results of an xe2x80x9cexposedxe2x80x9d scan. A xe2x80x9cdarkxe2x80x9d scan is a reading done without x-ray. A xe2x80x9cdarkxe2x80x9d scan simply activates the FETs on the x-ray detector panel. Thus, a xe2x80x9cdarkxe2x80x9d scan may determine the charge retention characteristics exhibited by the FETs activated to read the x-ray detector. By subtracting the xe2x80x9cdarkxe2x80x9d scan from the actual xe2x80x9cexposedxe2x80x9d scan of a desired object, the charge retention effects may be eliminated.
During an x-ray exposure, a similar phenomenon occurs whereby charge is generated in the FET as a result of the FET photo-conductive characteristics. When the FETs are turned off at the end of the exposure, the additional charge also leaks out and adds to the read signal in a manner analogous to charge retention. However, the additional charge cannot be removed because the additional charge resulting from the FET photo-conductive characteristics relates to the x-rays bombarding the x-ray detector. Thus, the additional charge resulting from the FET photo-conductive characteristics is not predictable or nor is it reproducible in a xe2x80x9cdarkxe2x80x9d image where no x-rays are transmitted. The number of FETs that photo-conduct and the amount of charge conducted by the FETs are dependent upon the amount of x-ray exposure and the object imaged, as well as upon the individual properties of each FET. Since a solid state x-ray detector is structured along rows (scan lines) and columns (data lines), the excess charge in the FETs may result in structured image artifacts or offsets which cannot be corrected by contrasting the xe2x80x9cexposedxe2x80x9d image with a xe2x80x9cdarkxe2x80x9d image.
Photo-conductivity is not as structured as charge retention. First, when a FET in the x-ray detector is turned on to be read, the FET is always turned on with the same voltage. With the photo-conductive effect, the xe2x80x9camountxe2x80x9d that the FET is turned on is determined by the intensity of the light reaching a given FET. The light reaching the FETs may vary among a wide range of intensities for all of the FETs on the x-ray detector. Second, regardless of how strongly each FET is affected by photo-conductivity (due to the light intensity at each FET), all of the FETs will be affected simultaneously. Charge retention induced by reading only stimulates one FET in any given column at a time. Therefore, photo-conductivity is much more unpredictable and is uncorrectable by a simple image subtraction method.
As noted above, the characteristics of digital image detectors inherently vary. Although there is a need to provide consistent and accurate image quality (and in particular, image resolution) within and across multiple medical diagnostic imaging systems, in the past there has been no automated technique for providing such consistency.
Thus, a need exists for a method and apparatus for correcting the offset induced by Field Effect Transistor photo-conductive effects in a solid state x-ray detector.
A preferred embodiment of the present invention provides a method and apparatus for correcting the offset induced by Field Effect Transistor (FET) photo-conductive effects in a solid state X-ray detector. The method and apparatus include reading out twice as many rows (scan lines) as actually exist in the X-ray detector. The additional rows may be read out between the actuation of xe2x80x9crealxe2x80x9d scan lines on the X-ray detector. The additional row readings may be used to measure the xe2x80x9csignalxe2x80x9d induced by FET photo-conductivity. In a preferred embodiment, the xe2x80x9crealxe2x80x9d rows may be read out during odd numbered scan lines, and even number scan lines will be used to measure the signal induced by FET photo-conductivity. To correct for the offset induced by photo-conductive FETs, an even row signal may be subtracted from the preceding odd row signal. The correction for the offset induced by photo-conductive FETs may occur after normal offset correction. In effect, the actual signal contribution from FET photo-conductivity can be measured independently on each column, for each row, and subtracted.